lerobot
lerobot copied to clipboard
Adding DOT policy
What this does
This PR adds the new dot policy. before merging, i’ll compare it against existing policies to make sure everything aligns 🤗! suggested by @Cadene
Hi! What is the plan for this PR? I can help if needed.
Hello @jadechoghari, is this still a draft?
Hey, I forgot to reply publicly! DOT is ready (thanks @IliaLarchenko), but there's some work around the Lerobot training script indirectly related to DOT, which @Cadene suggested. Additionally, I'm also training the policy from scratch see: https://huggingface.co/jadechoghari/ckpt-policy-dpk/tree/main on different tasks and benchmarking them, will also do it on different decoders. :) The PR is in a very good state though, and I'll open it soon!
This PR has been automatically marked as stale because it has not had recent activity (6 months). It will be closed if no further activity occurs. Thank you for your contributions.
This PR was closed because it has been stalled for 21 days with no activity. Feel free to reopen if is still relevant, or to ping a collaborator if you have any questions.